Yes, it's a strange title but I saw a pig in the rocks while working on this first render. Maybe it's not so noticeable now and other renders won't relate so much but I'm sticking with it.
Thanks to Volker for his massive enthusiasm with graph techniques.
Stars were obviously postworked and there were some minor level and colour adjustments.
There is another render I will link to tomorrow. I think it's much better than this one. More unusual and dramatic.
http://www.renderosity.com/mod/gallery/media/folder_156/file_1550911.jpg
spider pig, eh? ;D
really nice work
looks like a angel pig, with wings you know.
You are brilliant, sir. Great work!
cooool !
Very nice :). Your work with Voronoi has produced amazing results.
Very nice but I don't see the pig :)
Thanks for your comments. I'm eagerly waiting for when Renderosity allows me to post the next one. It's my favourite out of all the TG2 images I have created so far. I will link here.
It's my favourite because it finally brings together all the things I want to use in TG2. The fantastic gritty realistic nature of the surfaces, the atmospheres and the Global illumination but still I can mess with slightly surreal rock forms.
In the image or at least early in this planets creation the rock looked like an animal with it's snout pointing upwards. In particular it looked like a pig to me. I always see these shapes. I think it's because in Mojo you especially get these kinds of organic forms so I'm used to seeing it.
I hope some of you guys are experimenting with some things on that voronoi thread, if you can work some of it out in our ramblings.
Aside note - my Mac is spell checking me again. In the UK we use favourite not favorite. Color is Colour in UK. Meter is metre and there are many others. In case US people here think our spelling is all wrong. I don't know if my Mac can be set to change this. Any Mac users here know about this?
Here's the latest from the planet:
http://www.renderosity.com/mod/gallery/media/folder_156/file_1551512.jpg
This is...awesome.
cool...
Breathtaking POV.
Awesome work.
This look great, Efflux :).
As for the spellchecking, my best recommendation would be to turn it off - I did a bit of digging and couldn't find anything that localizes to the UK.
The last image is awesome. I can see a face in the foreground structure.
Thanks.
Ah yes. You saw it. Like an alien head?
If any of you have ever seen Slav's Mojoworld work you will often see heads and faces and creatures of all sorts coming out the rocks. His work is great and inspires me a lot.
I really like both pictures, the second really fascinates me! Good work!
Thanks.
I tweaked the planet a bit. You will notice the second render is warmer in colour. This is not just because of the low sun but also because I altered the sun colour. It looks cooler in high sun as well. Also, I changed the green areas a bit. It's begging for water now because it has almost beach like areas. I have some more small test renders that are worth rendering to full so you may be sick of seeing this planet soon. The voronoi work seems to have created a planet with endless variety and POVs. The techniques on the voronoi thread are a goldmine.
Not sick. More, please.
I agree about the goldmine.
To tell you the truth I've got sick looking at it. I have some other POVs like this:
(http://img379.imageshack.us/img379/6952/pigmooncc0.jpg)
I like the wide format and was going to render this out bigger but I'm bored with this planet. I could render this out big but I'm thinking my next step will be to bring the towers back but with more natural looking settings and use the more interesting surfaces and more natural environments. Like rock sculptures.
I should also show some WIPs of the Sine Clones planet although it's out of context with this thread. It's now abandoned. It got out of hand. This was meant to be the wildest thing ever but got horrendously complicated with about 500 nodes. These shots show some of the elements:
http://img263.imageshack.us/img263/7286/sinesclones1hf4.jpg
http://img522.imageshack.us/img522/4923/sinesclones2ve1.jpg
I don't want to make to much of this now, starting a new thread or anything. The Sine Clones project is ditched but I may resurrect some elements of it.
Actually, I've stopped with TG2 at the moment. My ideas are just future things to do.
HAHAHAHAHAHAHAHA!!! :D
500 nodes and this looks awesome! LOL Cool.
There were so many nodes with internal networks that when I opened a right click menu in the graph to add nodes (which you need if you're using internal networks) my PC just crashed. My Mac carried on but I got sick of it. Way too ambitious. I didn't show anything from this before because I wanted the impact of a first finished render. The techniques are all based around the things I described in my Planet Spinner post about how I used sines in that planet and the spires thread. Except the majority of the work was being done with sine nodes. It's nothing new and I used a load of those colour gradient hacks but the complexity got absurd. The thing is, some of the techniques discussed in the voronoi thread could be useful for this as well. I need to start it fresh some time in the future.
I can feel the craving. ;D
Anything you want to let us learn from you is fine with me. I'm still experimenting with the voronoi ideas you and Volker gave us. So this is great, as far as I'm concerned.
Quote from: efflux on November 01, 2007, 10:01:07 AM
I need to start it fresh some time in the future.
The beauty is that even when you abandon something it's all procedural anyway so you just recycle it all. It doesn't matter. You just change a few surfaces and you're onto a new planet. No time is really wasted.
I just tried hooking altitude into a perlin node and you can drive smooth steps in the terrain this way. Just the same as you can in Mojo. This has huge implications. Now I'm thinking of functions to control the stepping or using voronoi. It just goes on and on. This app is fatal when you have other stuff to do. Every time you open it, even to just render something you find something new.
Check this out:
(http://img511.imageshack.us/img511/8528/altitudeperlinvd7.jpg)
Like I said, altitude driven perlin. We can now do to the strata and outcrops shader what we did to the fake stone shader but ten times over. With some distortions and maths functions similar to what we used to destroy the fake stones shader on the voronoi thread, we can now take strata effects into the stratosphere. The strata and outcrops maybe uses voronoi but whatever it uses, those sharp edges are not always desirable.
TG2 is near complete Mojofication now, at least as far as it can go. Shame we don't have more basis functions.
This'll be hell with those shadows, if we try to use extremes like this but my idea is to blend this stuff with TG2's realistic aspects.
Some scale distortions:
(http://img210.imageshack.us/img210/236/perlindistortedstrataoi5.jpg)
This is what's known as strata and outcrops on acid.
This is the wildest one yet folks. Forget voronoi or spire threads. They haven't a look in on this one.
He he, just warming up. Insane possibilities with this one. No more time though.
Here's a graph:
(http://img119.imageshack.us/img119/1723/lateralgraphpv1.jpg)
The displacement shader is set to lateral only displacement direction. Plug this into a surface layers child layer. Play with the constant scalar value and displacement multiplier. You should get the results from the first image.
I don't think this was explained on the voronoi thread, but to distort scale using a power fractal, i.e. plugging it into scale on the Perlin, you are also using colour from the power fractal so you can just ignore the 0-1 colour setting. You can set this to higher values and not have it clamped. You'll just have to play with that to see what happens.
again, very cool and on the edge stuff, keep it up
I really like the second one. Special stuff you got here ;)
Quote from: efflux on November 01, 2007, 09:31:59 AM
http://img522.imageshack.us/img522/4923/sinesclones2ve1.jpg
are they sex toys ? ^^
Check out Armand's Sine Warriors Planet:
http://calyxa.pandromeda.com/otp/swotp.jpg
very nice....
Here's a first semi realistic environment where I tested this perlin strata thing:
(http://img208.imageshack.us/img208/3590/perlinstratapr5.jpg)
Efflux - I've just tried your setup from above without any results. I followed everything you said to a T by going by your graph and instruction. No reaction from the surface whatsover.
[EDIT] - NEVER mind. I wasn't in the right area of the planet and / or I didn't have the settings balanced correctly. While I have work to do this is what I have been able to do so far - [/EDIT]
That looks pretty darn cool, Calico.
That look good Calico.
I'm trying to ease you into understanding this. First you want to get those perfectly lateral strata like around those bluish towers I did. If I supply a tgc then it's not so good in my opinion and this technique is dead simple anyway. Once you know how to set it up then you have huge power to use it whoever you like. The real diificulty is actually simply in getting the right scales for the perlin and the displacement so it's the size you want to see in your scene.
The distortion is another matter but something that needs worked on and explained a bit better because for strata we definitely want that distortion. This is the real power behind this idea. It wasn't explained on the voronoi thread but it was in a WIP phase and still is. I'm still trying to work though it myself though. It's tricky due to TG2's separate use of colour, scalar and vector. However this does give TG2 flexibility.
My opinion is that this perlin strata is a major technique allowing for amazing forms that are very useful for natural effects. Bring in colour and it starts to get very interesting.
I had intended to stop with TG2 for a while but I will have to work through this idea a bit now. It's like the last frontier big frontier.
This should be a new thread really but never mind. These discussions are a bit messy anyway. Maybe in the future it can all be sorted out into better to understand tutorials but believe me, get into this and at least try to get some of it understood because the TG 0.9 users who aren't exploring these things are going to be left way way behind.
I think the displacements disappear when you go beneath an altitude of zero but that's a side issue. We'll get all this sorted.
You can of course get similar effects without using blue function nodes and using other shader setups but you don't get the same flexibility. I think that has been demonstrated on the voronoi thread. Can you do that stuff with fake stones (which are voronoi)?
This is how to set up the distortion fractal plugged into the perlin. The scale settings page effects the amount of distortion. So if you want not much strata distortion then make the scales very large and also here it's best to have smallest scale scale fairly large. The colour settings effect the size of the strata. You want these setting to reflect the amount of detail you want in the strata. Colour is irrelevant (at this stage). Ignore the 0-1 colour numbers. Switch off the clamping and adjust high and low colour according to the scales you want. You don't want lots of detail with this so make both high and low fairly large or the size of the strata, not 0-1 or higher than 0 for low - it all depends on your scales in the planet, you'll find the right settings (there are lots of ways but this is how to start simple - just adjust colour). Later displacements and detail in your terrain will add finer detail but there are no hard rules here except you don't want a displacement mess. The colour roughness will also straighten out the strata. Play with this stuff and you'll see.
A lot of these settings all interact with each other but you'll learn that. Some of the different settings will get you the same visual result so not so easy to explain it all. By experimenting you'll see.
Here's my first attempt at the "piggy technique". As you say Efflux the scale settings are crucial. But once you get it right it's very easy to manipulate. Lots of fun to be had here!
I've got a couple of others rendering at the moment and will post later.
Thanks for this Efflux - we'd be lost without you. Although I'm partly lost because of you! ;D
Regards
Kevin
Here's another variation. In this graph I retained the clamped 0-1 colour values but multiplied the fractal's output before going to the perlin. By retaining the 0-1 colour value it can be useful for other purposes. In this case I drove the blending shader on a power fractal which is supplying colour to the surface layer. You could also get outputs from after the perlin. There are numerous ways to drive the colours. In fact this is probably even more useful than using it for displacements.
The problem is to do with TG2's shadows. You'll find they screw up with displacements like these as they do with all distorted displacements. A serious problem but maybe it will be improved.
(http://img413.imageshack.us/img413/1533/perlinstratacolourwd8.jpg)
(http://img110.imageshack.us/img110/5575/perlinstratacolourgraphqt7.jpg)
Efflux, I just woke up to your explanations. Thanks. I'll have some time today to play with this and maybe mix it up a bit with voronoi discoveries I've made. I'll post what I find here and try to keep relevant. :D
Yes, the important thing is simply getting the scales of everything just right or things look rough and broken. Your image is cool and this is really easy so anybody can use it but try adding the distortion fractal if you can work out what I've said about how to do it. Then you get into even more scaling issues. the only way is to experiment.
Also, it might be a good idea to not go too crazy with the displacement multiplier value especially since we might run into shadow problems. My examples for everything are really extreme but I do that to really show what is happening.
There will be other issues and loads of tangents coming off this idea but I'll not go much further with it. The basics are laid out now. The distortion thing is very cool because that can be used for other techniques not just this strata thing.
Just to back up what I've said. Don't go to crazy with these things and try to make settings so that you don't get a lot of small rough details all over the place otherwise it won't look smooth and realistic. The terrain provides detail and further overlying displacements can provide detail. Some displacements can have low detail and be very smooth, just providing a general shape, otherwise you'll get clashes of displacement detail everywhere.
For example in the tests you guy's have posted, the terrain is already rough. Make the displacements have low detail. High smallest scale, low colour roughness etc. You'll find all this out by tweaking. My original examples are smooth but your terrains will provide roughness. It goes further though because really what you want is various layers of subtle things going on. A reasonably detailed terrain, other displacements to provide shape and then more surfaces over the top to provide the real detail. At least that's the way I work. It's not the way everyone does things but it's what TG2 can provide over TG 0.9.
Your pictures demonstrate the shadows problem by the way. This is one reason to not take this too far and have lots of huge rough displacements. Matt may improve that. I don't know.
Would someone do me a favor? I did this at home, but can't do it here. The Constant Scalar and Displacement need to have a balance. If you can just give me the percentages that you find are working, I can try it here. I'm getting nothing again.
Quote from: calico on November 02, 2007, 12:10:20 PM
Would someone do me a favor? I did this at home, but can't do it here. The Constant Scalar and Displacement need to have a balance. If you can just give me the percentages that you find are working, I can try it here. I'm getting nothing again.
call me crazy but the pic I did earlier had constant at 95, and dispalcement at -2000! also my terrain scale was 25000 lead in and feature scale 1080.
Kevin
Thanks Kevin. I'll try these.
Once I'm home again, I'll check mine. I believe they were all positive integers.
Too bad that I am out off time ... but my fascination carries on!
I'm out of time as well, Volker. I can't take this much further but the basics are here and I think this has massive potential. Mostly because any kind of interesting strata or altitude driven anything gives the scenes a great sense of height and scale.
I can try to answer any questions but I'm not starting to experiment further. Believe me though, this could go far.
Calico, are you trying to use the graph where I have multiplied the power fractal output since the other one doesn't have a constant scalar? The power fractal connected to the multiplier still has a 0-1 colour value and is clamped. The clamping keeps it from going over 0-1 (this is why in my previous graph where I ignored the proper 0-1 colour values, I had the clamping turned off - that was the quick way to hook everything up). The multiplying constant scalar value simply alters the scale to suit what size you want your strata to be. For example I used a value of 40 here. My terrain had a displacement amplitiude value of 1000 but also had a spike limiting and a displacement offset of 1000. I may have set the offset to that because the strata disappeared below 0 altitude. I can't remember but you could check that one out. My terrain was reaching a height of around 2000m where I applied that coloured strata - there are many factors which influence what your tallest peaks may be, not just the displacement amplitude. That last render was not much short of 2000m tall in what you are seeing. The distorting powerfractal was multiplied by 40 (the strata size) so do you start to see how these figures make sense?
You'll find the right balance of scales and once you do this you will be off and running with many possibilities. Just remember that we don't want tiny smallest features values on anything because we'll get messy displacements i.e. smallest scale or roughness values. This idea is to provide rough strata forms not infinite fractal detail. If you saw child@play's renders with the furry effect, where he tried this altitude thing. This is why he got that. Lots of fractal detail which caused lots of tiny displacements.
A lot of this stuff is very frustrating to get to grips and will take some time with but once you get it it will be easy.
Also, a side note. Although TG2 has nodes for scalar and colour. It doesn't matter in many cases. TG2 will know if you are using colour or scalar.
As I said, your scale values on the powerfractal influence the distortion. For more natural effects you don't want too much distortion. Set these values really high. Several times your scale values on your terrain. This is because if they are low they will distort your strata within too small a distance and really swirl up the shapes, unless that is what you want.
efflux - Thanks so much for the explanation. I've simply kept the nodes from the very first graph with nothing but the Perlin noise with the Constant Scalar and Displacement. Understanding is slowly developing about how this all works with the Perlin. I even tried mixing the Perlin terrain with a Voronoi terrain and found a lot of displacment messes. It's interesting to hear your point of view on the small fractals. Focusing on the displacements without so many fractals is sort of a leap in another direction I have yet to explore. Last night I was just decreasing the values of my displacement and constant scalar, but keeping the ratio about the same and found it doesn't really matter that I slapped it out there so large as before. This program can deal with numbers in a way I have yet to discover.
What this is doing is increasing my overall grasp of Terragen 2, which is what I hope for everyone in the future. We're so privileged to have so many knowledgable and helpful people here.
This product is just one big adventure.
Keeping the displacement multiplier and the constant scalar about the same is fine. In my scene my displacement multiplier is 10 so less than the constant scalar of 40 because I don't want too much displacement but yes these values should be close. It all depends on what effect you want but really you don't want small strata size (the constant scalar) and big displacement.
The only reason I know this stuff is from using Mojo and I understand it's not that easy because it took me a long time in Mojo but you have one big advantage in TG2. It is much much easier to use and learn. With Mojo many people shared their finished planet files. These were often fantastic pieces of work but it didn't help people understand how they were built in my opinion hence now you see less cool renders in the gallery because some of these Mojo masters are not working much anymore.
It's the concept you have to grasp. Your terrain may have a lot of detail. What you want to do with the strata is push the terrain out in stratified bulges but you don't want the strata having lots of detail. It is being controlled with altitude which means that lots of detail in the strata will create lots of fine strata which will result in an effect where there are so many small strata that you just see a lot of noise. You will have seen this no doubt in your experiments. Once you grasp the idea, then the settings will make sense.
The problem is knowing the best way to get people learning. We definitely need all this stuff put into easier to understand tutorials.
Volker and I supplied files on the spires thread. I saw a few people using Volker's file but I saw nobody really tweaking it to create something new which suggests they didn't learn exactly how Volker got the results otherwise they would have experimented a bit more.
I'm not working in any new explorations which takes up too much time but what's already done I have a little time to help with. From tomorrow onwards I won't be producing any new work for a while.
On these various threads we have four basic and very powerful techniques. Clamping, masking, altitude driving and distorting. Four fundamentals that give you immense power in TG2. We need it documented properly but this takes time. The distortion is a new thing that I've only just completely mastered.
This is something I was working on last night (and just now re-rendered after some more tweaking) and like it, except for some strange shadow effects. Nevertheless, with some level tweaking of lighting in PS, I like this end result.
This is nothing more than taking your original Perlin graph, efflux, and tweaking it as well as placing it with some other displacements I've worked on before such as Fake Stones. But, the Perlin is the major player in this one.
I'm not sure you saw what I did here - http://forums.planetside.co.uk/index.php?topic=2662.0 , but this was completely taken from that thread about voronoi. I created it using the illustration about what you all were doing and tried to follow along. While I readily admit I didn't follow it all closely, I learned a lot and the proof (as someone says) is in the pudding. I didn't use a TGC from anyone for this work, though.
Usually, when I get a TGC from here, I dissect it and learn all I can. I ask questions. It's too bad we don't all have time for this or that some might actually be just simply too greedy to participate and take our work up like ravenous starlings at the beach. I sometimes only have time to learn the generalities, especially with large lessons from BigBen or Volker or you, efflux. Patience, though, will win out in the end. All I need to do is come back here later and dig in again.
I agree that some clear tutorials would help immensely. It almost shocks me that in the software world of 3D art that learning the tool takes so much energy, when being creative should be the goal. Whenever the tool gets in the way of creativity, it must never be a good thing for the end products created.
Quote from: efflux on November 03, 2007, 09:21:15 AM
The problem is knowing the best way to get people learning. We definitely need all this stuff put into easier to understand tutorials.
Volker and I supplied files on the spires thread. I saw a few people using Volker's file but I saw nobody really tweaking it to create something new which suggests they didn't learn exactly how Volker got the results otherwise they would have experimented a bit more.
I'm not working in any new explorations which takes up too much time but what's already done I have a little time to help with. From tomorrow onwards I won't be producing any new work for a while.
Quote from: efflux on November 02, 2007, 07:44:55 AM
Here's another variation. In this graph I retained the clamped 0-1 colour values but multiplied the fractal's output before going to the perlin. By retaining the 0-1 colour value it can be useful for other purposes. In this case I drove the blending shader on a power fractal which is supplying colour to the surface layer. You could also get outputs from after the perlin. There are numerous ways to drive the colours. In fact this is probably even more useful than using it for displacements.
The problem is to do with TG2's shadows. You'll find they screw up with displacements like these as they do with all distorted displacements. A serious problem but maybe it will be improved.
(http://img413.imageshack.us/img413/1533/perlinstratacolourwd8.jpg)
(http://img110.imageshack.us/img110/5575/perlinstratacolourgraphqt7.jpg)
I am trying to get the result like yours (not color but strata displacement)...but its not working....what am I doing wrong???
OK, I will take a look at your file, dhavalmistry. If this all becomes too much of a problem I'll put a tgc on here but I'm not a big fan of that especially when I see them simply being used as is without any tweaking or development which shows some understanding. Not that you guy's will do this but some will with no advance in understanding which is the aim of discussing this stuff here. If even one person susses out these techniques properly then it's a success.
Calico, TG2 is actually quite difficult to learn. After all, we are creating entire virtual worlds here so it is not easy to learn but TG2 is very easy in comparison to other similar apps which are so poor in their UI, I don't want to use them again.
Quote from: efflux on November 04, 2007, 06:32:04 AM
OK, I will take a look at your file, dhavalmistry. If this all becomes too much of a problem I'll put a tgc on here but I'm not a big fan of that especially when I see them simply being used as is without any tweaking or development which shows some understanding. Not that you guy's will do this but some will with no advance in understanding which is the aim of discussing this stuff here. If even one person susses out these techniques properly then it's a success.
Calico, TG2 is actually quite difficult to learn. After all, we are creating entire virtual worlds here so it is not easy to learn but TG2 is very easy in comparison to other similar apps which are so poor in their UI, I don't want to use them again.
hi guys,
efflux, i agree with you, sometimes hints can do more for people to understand than a given tgd or tgc. but on the other hand, sometimes it's easier for people to learn when they can see the base structure (myself for example ;)). i think sharing network views/setups can be a great help when you're learning new, abstract things, especially when you're not an expert on e.g. maths.
hopefully we'll be able to do a tutorial of these things soon, containing both, information and maybe some very basic node setups.
enough hijacking for now, keep up your great work everyone
cheers...
Hi child@play,
You played with the ideas and got some results which is cool. By exploring the settings a bit you can work out how it works but I know it isn't easy. I'm going to put a file on here which is a tweak of dhavalmistry's and then it can be seen how to set this up but dhavalmistry put a graph together so that's cool. I'm happy to tweak it now and everyone can take a look. My colour manipulation on my last graph was a bit quirky and odd to be honest. Not too helpful. I'll have to explain the oddities of that but it's all a work in progress.
well, i think the main idea in the way i'm learning is to get an insight when your a total stranger to that subject. when you check the nodes closely, how the perling looks after doing this or that, you can easier see what can be done to get what you're looking for. you easier get the feeling of how the output will look when you multiply/divide/whatever that particular node by this or that value/scalar etc. 'ah, the noise looks this way, now i have to do that to get closer to what i want' (at least my brain keeps telling me ;)).
that's just the way i'm doing things, sorry for getting off-topic for that crap, lol
OK, dhavalmistry. Here is a quick change to your file so you can see a suitable use of this technique. Lets forget the multiplying scalar. My colour method in the last graph will just confuse things further and was not a helpful step at this stage so lets get back to simply changing the fractal's colour values. The clamping does not clamp from 0-1 but simply clamps the values you enter so don't worry about the clamping. Explanations I gave before may have been confused about that issue. Some changes I made:
I increased your distorting power fractal's scales so that the distortion is less but you can still see some distortions going on. I also increased the smallest scale because there is no need for small scale calculations in the fractal, we only want largish features. I also changed it from perlin ridges to perlin billows. Billows are a bit smoother but that change was not crucial. I reduced the colour roughness - play with this one - it has a lot of effect. You'll see. Also, be careful with noise settings and warping. I have reduced these. They will distort up your strata even more. I changed the displacement multiplier to 10 not -10. In this scene, 10 provides quite high displacement - bear in mind we are pushing displacements around a lot which distorts geometry and there is this shadows problem so lower might be better. The quick render settings in this file will demonstrate this - black patches. It is a ray traced shadow problem. Matt has mentioned something about that. It may be improved in future versions. Another help when distorting things like this is to increase the detail settings to higher than usual when rendering. You may notice the geometry around the strata ridges gets a bit rough. This strata technique is actually very useful for simply colour never mind displacing - you should try that. The file is not perfect and I'd suggest keeping subtle with the displacement multiplier but there are so many variables that effect the problems. If your strata are thicker then the geometry is not bent so much so you can have bigger displacements. This is what it boils down to. TG2 is building everything up with triangles. The more that its pushed and bent the more likely you'll see that geometry rather than a smooth shape.
I removed the surface layer fractal breakup. The simpler and clearer this is the better.
The other point is that you used colour from the distorting fractal. It's not really achieving much because of it's distorting type settings. I have set up a blend from two fractal to get colours that will follow the strata. A simple way to get colour and more in line with my earlier examples rather than my later more quirky colour set up that will simply confuse things.
I've turned the GI to 0. I never use that until final render. It's just so anyone looking at this file can simply render it quickly to see the result.
I edited the file again. I suddenly realized there was not lateral displacement with a compute normal. This is ideally what we want to do here - have strata appearing to displace more on steeper slopes. We just need lateral displacement and I've decreased the displacement amplitude.
thanx efflux....I will check it out!...and thanx for the detailed explanation...
hi,
looked up both ps and ps tweaked tgd, and i see what you're doing, but maybe i got wrong what you're looking after. maybe something like this? ??? ?
What did you do to the files?
The difference is that the first one displaces outwards from the normal i.e. could be in any direction. The second file only displaces outwards horizontally and slightly less because I adjusted the displacement amplitude. It will displace this way even on flatter ground but will give the impression that most of the displacement is happening on the hillsides. For strata we really just want this horizontal or lateral displacement. You need to hook a compute normal after the displacement to do this.
ah, forget it, i'm just going bonkers over here again. i better shut up now
hey efluxx....how do I get the strata wave effect and not go in straight line??..
Reduce the first two settings in the scale section of the power fractal attached to the perlin node or increase that fractal's colour roughness which will also change this. Try that first. Change it to 2 and you'll see a slight difference. Keep the smallest feature size in the scale settings the same though. All these settings are interlinked in various ways.
I deleted my message just posted in this space. I'm confusing myself and you.
After setting the colour roughness to 2, try reducing just the feature scale setting by half. You'll see it becoming even more swirly.
The reason for more swirls with smaller feature scales is that the feature scale is moving the shapes about. Once it becomes smaller it moves those shapes in a smaller space but it has to be quite large or you'll get incredibly tight swirling effects. This is why it ideally needs to be several times the size of your terrain feature scale for more subtle natural results.
ah it works...thanx efflux :)
In one sense TG2 is a little more confusing than Mojo. Mojo does not have separate colour. All your control is in it's scale settings but in TG2 you have some separate colour control. This does actually make the concept a little more confusing but it's quite versatile.
I just edited a confusing post above so read what I've added there now.
Efflux I think that the perlin noise function must have a vector as input, to have any influence. Pressing The noise function twice only shows a grey mini previev.
We're not really using it as intended but then again TG2 allows what we are doing so we can hook things up in unexpected ways. We are only using the perlin to attain a profile for the strata and all we need is an altitude value for position but the preview window will expect a vector I guess and as is can't really display the effects of the perlin. You can hook in other functions such as voronoi instead of perlin or attach a scalar to change the noise or do as I did (in the initial planet on this thread) and hook in a get position in geometry. Then you'll get a preview but of course you don't have the strata effect. If you change the perlin 3D scalar for a voronoi 3D scalar then you'll see the strata steps attain a voronoi profile rather than perlin. You'll see it's sharper edges. You could use a bunch of maths functions like we did on the voronoi thread to alter the strata profile.
The ideal thing would be to have a changeable positioning input or a choice within the UI of the power fractal then we could do this easily with just one powerfractal node or at least get the strata that way. We'd still need a distorting fractal.
The voronoi can be quite cool actually. Anyone here experimenting can try what I described above. Change the perlin 3D scalar for a voronoi 3D scalar and also try plugging a get position in geometry into the perlin (or voronoi) instead of altitude then you'll have the basic set up for achieving what I did on the initial planet.
I rendered that smaller image previewed earlier to bigger size. 4000 x 500. There were some minor tweaks so it looks a little different. This will probably be my last image for a while unless I just render more from this planet but no new planets.
http://excalibur.renderosity.com/mod/gallery/media/folder_156/file_1556551.jpg
OMG
I'm so envious of your knowledge. Brilliant work!